Background:Systemic Lupus Erythematosus (SLE) can first present with severe or critical disease leading to hospitalization. Prompt recognition of the disease in hospitalized patients may lead to early institution of treatment and improve outcomes. We have recently developed a clinician-friendly algorithm for SLE diagnosis based on classical clinical and serological SLE features [SLE Risk Probability Index (SLERPI)]1.Objectives:To determine the clinical phenotype of SLE patients first diagnosed during hospitalization, the interval between hospitalization and SLE diagnosis and the potential impact of SLERPI on early diagnosis.Methods:Mixed prospective (from June 2020 to January 2021) and retrospective study of SLE patients from “Attikon” cohort (n=820)2. Clinical phenotype was divided into 10 core domains (neuropsychiatric, thrombosis, nephritis, serosal, haematologic, pulmonary, cardiovascular, gastrointestinal, skin-joints, other). Chart review and patient interview was performed to assess the lag time between 1) the onset of symptoms and 2) the hospitalization and the final diagnosis. Demographic and clinical characteristics, SLERPI and SLICC damage index were recorded for each patient at the time of diagnosis. SLE diagnosis was based on at least one of the three existing classification criteria.Results:Out of 820 SLE patients, 202 (24.6%) diagnosed during hospitalization were included. Among them, 185 patients (91.5%) were hospitalized because of a lupus related feature, while in the remaining 17 SLE patients, hospitalization was due to non-lupus related manifestations. The most common lupus-related clinical phenotype leading to hospital admission was neuropsychiatric lupus (n=51, 25.2%) with cerebrovascular events constituting the dominant clinical syndrome (n=8/51). Thrombotic events (n=32, 15.8%), mainly pulmonary embolism (n=20/32), cytopenias (n=32, 15.8%), lupus nephritis (n=30, 14.8%), skin-joint disease (n=26, 12.8%) and serositis (n=24, 11.8%) were also common as dominant manifestations. Pulmonary disease (n=16, 7.9%), heart disease (n= 4, 1.9%) and gastrointestinal disease (n=2, 0.9%) were less common. On admission, 11.3% of patients (n=23) had symptoms from at least 2 clinical domains as defined. Most patients (93.5%) had multisystem disease while only 6.5% had organ-dominant disease. Early diagnosis (within 3 months from hospitalization) was established in 86.6% while 27 patients had their SLE diagnosis more than 3 months from hospitalization. The mean lag time between the hospitalization and the diagnosis was approximately 14 months (SD 19.9). Overall, the mean interval between the onset of symptoms and the diagnosis was 48.2 months (SD 73.2). Importantly, a SLERPI >7 (suggesting probable SLE) at hospitalization was present in 92.5% of SLE patients with delayed diagnosis.Conclusion:One out of four SLE patients first present with moderate to severe disease necessitating hospitalization, while in approximately 15% of such patients, diagnosis is initially missed. Application of the SLERPI may facilitate early SLE diagnosis.References:[1]Adamichou C et al. Ann Rheum Dis. 2021; DOI: 10.1136/annrheumdis-2020-219069.[2]D Nikolopoulos et al. Lupus 2020; doi: 10.1177/0961203320908932.Acknowledgements:This project has received funding from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (grant agreement No 742390)Disclosure of Interests:None declared
Correspondence on ‘Lupus or not? SLE Risk Probability Index (SLERPI): a simple, clinician-friendly machine-learning-based model to assist the diagnosis of systemic lupus erythematosus’
